"Ft. Collins Flood"
28 July 1997

This case focuses on the flood which occurred in Fort Collins, Colorado. The data used for this were presented as a laboratory exercise during the Hydromet residence course held at COMET.

Materials:

1. Hand analyzed maps from 2100 UTC 28 July and 0000 UTC 29 July of surface isobars and dewpoints.
2. Maps of 700 mb dewpoints & plots, 500 mb heights, and 250 mb plots and isotachs, all at 0000 UTC 29 July.
3. Isohyet analysis of previous night's excessive rainfall centered mainly northwest of Fort Collins.

1. Review data through 0100 UTC 29 July 1997.

2. Now look at data through 0430 UTC and examine the Fort Collins storm complex. Make sure to examine satellite and radar data.

3. Might there be atypically high rainfall rates?

   • Look for warm rain signature such as relatively warm cloud tops and low storm centroid (precipitation concentrated in lower part of the storm).
   • What environmental conditions may be contributing to warm rain processes (consider both meteorological and geographical)?

    

4. Look at the radar derived precipitation products.

5. Consider three important contributions to flash flooding:

Information from the Colorado State University research radar (CHILL) for this event can be found on their website.

1. Storm elements were moving to the north-northeast throughout the northeastern part of Colorado. In the Fort Collins vicinity the storms began moving south to north repeatedly by 0300. This is likely due to the combination of regeneration westward toward the foothills and the steering current toward the north-northeast.
2. During the period of 0300-0430 UTC 29 July, the activity over the Fort Collins area became more of a quasi-stationary storm system with some tendency for elements to move northeastward. The lowest radar tilt from either KFTG or KCYS could not accurately depict the low-level flow supporting this system. However, the Colorado State University research Doppler radar was located much closer to the activity. This depicted strong outflow from the east and into Fort Collins that quite likely enhanced the low-level support for new cell generation against the foothills. Regeneration of storm cells appears to have balanced the movement of cells to the northeast. Eventually, this enhanced upslope flow ended and the storm system moved off to the northeast as it dissipated.
3. Conditions on 28-29 July showed unusually moist, tropical conditions for the region as seen in the atmospheric sounding (especially the exceptional precipitable water values), and the atypical low-level dewpoint temperatures. As the storms were in progress certain characteristics, particularly of the Fort Collins activity, were noteworthy from the perspective of precipitation efficiency:

• High reflectivity, but low centroid. A four-panel view of the lowest radar tilts from KFTG shows the low-centroid nature to this storm. The four panels clockwise from top left show data from radar tilt angles of 0.5^o, 1.5^o, 2.4^o, and 3.4^o. Much of the storm intensity (and thus the precipitation production) was below the radar beam at 2.4^o, or about 4.5 km (15,000 ft) AGL. With storm bases about 500 m AGL, a large area of the precipitation growth was in the above-freezing layer of the cloud. Warm-rain processes were enhanced, thus enhancing the precipitation efficiency.
• In support of the radar’s suggestion of an enhanced warm-rain process, note the relatively warm cloud-top temperatures and the lack of lightning activity when compared with nearby storms to the east.
• Finally, the terrain focus increases greatly in Fort Collins where the foothills of the Front Range rise abruptly immediately to the west of the city. This would serve to enhance the low-level upward motion in a cloud system already characterized by efficient precipitation growth in the low-levels.

4. Fortunately, excellent ground reports exist for this event. The radar-derived precipitation guidance appears to compare better for the tropical Z-R (Z=250R^1.2) than for the default 88D settings (Z=300R^1.4) in Fort Collins. But remember, the radar is not necessarily performing better just because a radar bin accumulation (~4 km2) matches a point estimate from the collocated gauge. It appears that the area of large accumulations with the tropical Z-R may be a little too large, but better than the area with the default Z-R.
5. Other precipitation extrema occurred that day that did not seem to exhibit the same magnitude of tropical precipitation processes. Although warm-rain processes appeared to have played a role in other precipitation extrema, the Fort Collins complex exhibited the most impressive precipitation efficiency due to the low level orographic lift. Other storms had colder cloud tops, more lightning and ice processes, and were not very well represented by the tropical Z-R. Radar parameters will not always be able to represent all storms occurring within the radar domain. Forecasters are faced with the difficult task of identifying and reacting to the signatures that make a particular storm unique. Signatures of exceptional precipitation efficiency have been seen in other major flood events in recent years and thus it's important to understand the physical processes affecting rainfall rates. The radar is only one of several important tools.
6. Rapid regeneration of echoes increased the duration of the heavy rainfall period over Fort Collins. An unusually warm tropical environment greatly increased the precipitation efficiency. Enhanced duration and intensity of rainfall combined with urban runoff complexities and resulted in a very major hydrologic response. However, the intense rates and the urban environment were even more important than the duration, as is often the case with urban flash floods.

Case Study 011